Product intent: new users get summaries of their newest videos fast while the
back-catalogue fills behind, within the shared rate gate. The background batch
currently processes in subscription/channel order, not newest-first — this spec
closes that gap (collect candidates, sort published_at DESC NULLS LAST, process
in order, gate unchanged). Also corrects the docs to describe Try-now as
onboarding prioritisation, explicitly removing the prior "looks organic to
YouTube" traffic-disguising framing — rate limiting is respected, not evaded.
homelab-integration.md gains a "Scheduled discovery" section documenting
TAPIR_DISCOVERY_INTERVAL and TAPIR_FETCH_RATE and the load-bearing
single-replica constraint (in-process scheduler → replicas: 1 is required;
>1 double-runs discovery). VISION Stage 0 carries a pointer to ADR-018's
gate-clock reset so nothing in docs implies the window started before
unprompted use was possible.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Stage-0 usability fix: tapir serve runs discovery for all users on an
interval (reusing the existing Runner.Loop, enumerate-users-then-withUser),
auto-summarize defaults ON for Future-B users so the list fills itself, and the
ADR-014 process-wide per-egress-IP rate gate is confirmed/finished in the same
slice because in-process + auto + multi-user makes it load-bearing. Records the
single-replica constraint as load-bearing, and a fallback (auto-summarize OFF
until the gate exists) so the dangerous combination never ships half-built.
Small tapir slice to make the gate measurable as written: append-only
login_events (RLS, per-user-per-day throttle) + a union query over reads
(login_events) and acts (summary_actions) for distinct-active-weeks. Carries the
honesty caveats (unprompted not measurable; data accrues from deploy; week-bucket
noise at low N) and the delete-cascade footgun (no FK, needs explicit delete +
test) from the prior delete work. Out of scope: analytics, prompt-tracking,
dashboards.
429 from the caption endpoint means the IP is rate-limited (retry later),
not that the video has no captions. Distinguishing it from SourceNone is the
prerequisite for the runner's backoff/retry logic.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Stage-0 ui-spec (ADR-011) predated multi-user and several UX
features. Appended a "Deviations and additions (as-built)" table —
without rewriting the spec — recording each feature shipped beyond it
(multi-user+RLS, registration gate, per-user YouTube connect, account
management, immediate web summarization, charmbracelet spinner,
auto/manual mode, public landing page) with the why and the
commit/ADR that covers each. Preserves the intent-vs-reality split.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The .feature spec lagged shipped behaviour. Added three files, no
duplication of existing scenarios:
- registration.feature: new Dex subject -> registration gate (users +
user_identities), returning subject straight through, account delete
is tapir-side only and leaves other users intact, clean
re-registration (ADR-012, ADR-013).
- summarize_mode.feature: auto summarizes every new video; manual
(default) leaves them unsummarized until queued; queued video is
processed and the flag cleared (migration 006).
- landing_page.feature: unauthenticated / -> /welcome, Get Started for
guests, summary link + logout for authed users, logout -> /welcome.
Scoped to built features only — no Vimeo/Whisper/billing scenarios.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The C4 view predated the web transport. Added a "Web surface" section
with a component diagram and prose covering: tapir serve (HTMX+Templ
over the unchanged store), oidc authenticate-only session, registration
gate (users + user_identities), per-user YouTube web connect callback,
account disconnect/delete (ADR-013 tapir-side only), immediate
summarization via background goroutine + HTMX status poll, and
auto/manual summarize mode (migration 006). Relabeled the L2 http node
to tapir serve. Corrected the out-of-scope isolation bullet: RLS is live
(ADR-012, migration 003), not deferred. ADR-003 stance preserved — the
engine/ports/sinks core is untouched; the web is a new transport.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The ER diagram and notes predated migrations 002-006. Brought them to
code truth:
- add summary_actions (002), user_identities (004), video_connections
(005) with real columns/constraints; rename token_secret_ref ->
token_ref to match migration 005.
- add users.auto_summarize and videos.summarize_requested (006).
- note FORCE RLS coverage and sink_deliveries' EXISTS-derived policy
(003); user_identities NOT RLS'd.
- mark AI_CREDENTIAL and SUBSCRIPTION as planned (no table exists; only
videos.subscription_id, no FK).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The "Isolation invariant" section said enforcement was dormant at
Stage 0. ADR-012 turned it on: Postgres RLS ENABLE+FORCE on every
user-owned table (migration 003_rls.up.sql), keyed off the
tapir.current_user_id GUC set by the store's withUser helper, deny-by-
default on an unset GUC. The Stage-2 two-user isolation test
(internal/adapters/store/rls_test.go) is pulled forward and passing.
Kept the ADR-011 single-user history honest.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Workstream A: public /welcome landing page (bubbletea aesthetic, one Dex login
flow, logged-in shortcuts) — with the oidc.go facts verified against main, incl.
the two corrections that only surface from reading the code (logout must redirect
to /welcome not /auth/login; bare-/ vs deep-link redirect split).
Workstream B: reconcile the guardrail docs against deployed reality (v0.4.0) —
auth.go comments, data-model isolation status + migrations 002-006 schema,
architecture web surface, use-case scenarios for the Stage-1 features, ADR
ordering, and a requirements-vs-shipped deviation check. Structured as two
parallel workstreams so the doc audit isn't done cursorily alongside the build.
Add GET /oauth/youtube/connect and /oauth/youtube/callback, mounted inside
the login + registration guard so CurrentUserID is always set and every
connection binds to the authenticated tapir user.
- connect: generate a per-user CSRF state (single-use, short TTL, bound to
the user), redirect to Google consent with access_type=offline and
prompt=consent so a refresh token comes back.
- callback: verify the state belongs to this user, exchange the code via the
existing auth.Exchange, persist the refresh token under a PER-USER ref
(web.YouTubeTokenRef = "youtube/<userID>/refresh_token") so tenants never
collide, then UpsertConnection (provider=youtube, status=active). Any
failure renders a clean error page and leaves no half-written state.
Reuses auth.Exchange and adds auth.AuthCodeURL (offline + consent) rather
than the CLI's listener/terminal flow (ADR-006: web flow, not CLI). The
ConnectHandler depends on a narrow web.Connections port, not the concrete
store. Wired in cmdServe only when YT client credentials are present;
TAPIR_YT_CONNECT_REDIRECT_URL configures the callback URL. Per-user token-ref
scheme documented in docs/homelab-integration.md.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
ADR-011's single-user authz (ID-token subject must equal AllowedSubject,
else 403) is replaced by ADR-012's model: Dex authentication is the only
gate — any Dex-authenticated subject may establish a session. Whether that
subject has a tapir user, and routing to registration if not, is decided
downstream in internal/web (next commit).
Removals (noted): oidc.Config.AllowedSubject + its required-field check + the
callback 403 branch; config.Config.AllowedSubject + TAPIR_ALLOWED_SUBJECT env
wiring; the AllowedSubject arg in cmdServe. ui-spec.md updated to reflect the
supersession. Sessions, cookie signing, login/callback/logout unchanged.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Light + dark list, mobile (375px) card reflow, and the reader detail page in
both schemes — captured via web-shot against the seeded serve. Visible evidence
that the Top 5 UX fixes landed.
Reviewer pass against running `tapir serve` UI, seeded with 11
representative summaries (5 channels, rich+sparse, local+fallback,
varied action states). Captured via Playwright on koala k3s across
desktop/mobile and light/dark color schemes.
Two blocking findings: dark mode is unreadable (summary text near-black
on a dark canvas — `color-scheme: light dark` declared but `--fg`
hardcoded and no body background), and the list is a 6-column table that
does not reflow on mobile. Plus a sub-4.5:1 muted color, an unstyled
"admin table" surface, and small correctness nits (stray `· — ·` meta
join, cramped fallback badge). HTMX action toggles verified working.
Includes UX-REVIEW.md (severity-tagged findings + Top-5 sleek list) and
16 screenshots. No code changed — drives the next UI iteration.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Reader + watch/skip/save (instruments the Stage-0 'acts on a summary' metric),
HTMX+Templ over the existing store, Dex OIDC login with single-user allowlist
authz (authn now, tenancy deferred), deployed at tapir.d-ma.be via Flux GitOps
with ESO secrets and in-cluster postgres18. Build decomposed into 4 gated lanes.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Vimeo texttracks API is owner-only (worse than YouTube, no public timedtext) → defer.
Whisper viable as a no-caption fallback; berget/whisper-large-v3 already on the gateway
(cloud), local options on iguana/koala carry setup+GPU-contention cost (ADR-007). Both
investigated inline (session sub-agents are network-sandboxed).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
main.go dispatches `tapir auth` (interactive OAuth → persist refresh token via
SecretStore) and `tapir run` (wire YouTube source + local summarizer + store
sink, build engine, run the dedup-aware loop). Config-driven so live creds plug
in at demo time; SIGINT stops the loop cleanly. Block kept minimal so Worker E's
list/show cases union cleanly at merge.
Add .env.example documenting every TAPIR_* var and a README demo runbook. Pin
the summarizer alias-as-config decision and record the max_tokens fix in
docs/homelab-integration.md (clears two `confirm` items).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Implement ports.Summarizer in internal/adapters/summarizer. It routes through a
local Primary endpoint first and an optional BYO Fallback, owning the routing
itself (not delegating to llm.Router) so it can record AIProvider, AIModel, and
FallbackUsed on domain.Summary. Prompt asks for JSON {summary, highlights,
takeaways}; the parser tolerates thinking-model fences/reasoning and rejects an
empty summary.
The summarizer is the single egress point for content toward an AI model, so it
enforces the local-first guarantee from ai_routing.feature: with no BYO
configured (nil fallback) there is no external endpoint, so content reaches the
local stack and nowhere else. Tests assert all four scenarios via a fake client.
Model alias is config (TAPIR_SUMMARIZER_MODEL, host/name) — not hardcoded;
docs/homelab-integration.md notes it stays `confirm` and that thinking models
need an explicit max_tokens or they return empty content.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Conductor verified against the live koala k3s cluster before launching the
adapter workers: LiteLLM is in ns ai-stack, ClusterIP 10.43.159.89:4000,
off-cluster NodePort 30401. The doc's 31234 was actually llama-swap. Also
flag that sk-local-123 now 401s — use LITELLM_MASTER_KEY from the vault.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Lists the egress the build assumes (Go module proxy, toolchain download,
raw.githubusercontent for golangci-lint, github for non-proxied modules) and the
runtime assumes (LiteLLM gateway, brain-mcp, YouTube/Vimeo APIs, per-user BYO-AI
hosts only on opt-in), so a locked-down koala act_runner or dev env knows what to
allow or which GOPROXY to set. Notes the claude.ai sandbox allowlist is separate
and unrelated.
Gherkin spec for connecting YouTube/Vimeo accounts and configuring optional
per-provider BYO AI credentials: connections sync subscriptions, tokens/keys are
stored only as secret references (never in the clear), revocation stops watching
but preserves history. Encodes the secrets-by-reference and data-isolation
guardrails (ADR-002, ADR-006, data-model).
Gherkin spec for the routing guardrail: local produces the summary by default;
on local failure, fall back only to a user-configured BYO provider; with no BYO,
queue for retry and never send content to a third-party model. Encodes the
local-first/user-owned principle (VISION) as executable behavior.
Gherkin spec for the core use case: captioned video is summarized and delivered;
no-transcript video is recorded as skipped; unsubscribed channels are ignored;
already-summarized videos are not reprocessed. These scenarios seed the use-case
test suite (Clean Architecture core tested through fake adapters).
Per-user-isolated entities (no global cross-tenant video table per the S5/grill
correction), secrets stored by reference only (ESO/1Password, never the token),
brain delivery modelled as a sink_delivery row rather than brain-specific tables.
fallback_used recorded per summary as the Stage 0 quality signal. Future C dedup
and sharding explicitly out of scope.
Context and container diagrams, sequence diagrams for the core summarize-new-video
use case and the local-first/BYO AI routing fallback, and the Clean Architecture
layering. Ports & adapters keep the engine provider- and sink-agnostic, making
standalone-vs-homelab a wiring choice (ADR-003), not two codebases.
mathias
mathias